This is a resubmission of an application for 5-year renewal of a family study of alcohol dependence (AD) that nests a high risk genetic/family study within an epidemiologic framework - a novel paradigm for research on AD. Its oversample of African-American (AfA) families (a neglected group in the AD literature)permits testing long-standing models of adolescent and young adult alcohol involvement that have been based for the most part on Caucasian ("Majority- "maj") youth. In the renewal, we propose continuing to assess at 2-year intervals offspring (70 percent 18 or younger at intake) in low and high risk families. Ascertained from birth records, families are classified as high risk (by maternal report that the father is an excessive drinker) or low risk (no such report) Additional families of men with 2+ DWI's (from driving records) enrich the high risk sample with severely alcoholic men. Mothers and offspring are administered a comprehensive psychiatric interview by telephone (fathers are given a shorter interview);offspring are re-interviewed at 2-yr intervals. During the first grant period, 472 families (255 AfA, 217 Maj) have been enrolled;the last intake sample will be enrolled by renewal month 15. Follow up rates for the first intake sample of offspring (n=221) were 90 percent (85 percent completed, 5 percent pending at this writing). Further follow-up is on hold pending funding. Our projections are that by the end of the renewal period, 3 assessment waves will be available on all offspring, and 4 waves on about half. Our aims in the first grant period were to test the deviance proneness and negative affect regulation models of AD vulnerability;to study sibling influences on alcohol and other substance involvement, and to investigate cross sectional and longitudinal influences of parenting, paternal AD, other parental psychopathology, offspring own psychopathology, traumatic events, and life transitions (marriage, work, school, parenthood) on alcohol (and other substance) use outcomes - in AfA and Maj families by risk. These aims are retained in the renewal, with an emphasis now on application of longitudinal data to test for interactions between family risk status and ethnicity on progression of alcohol use disorders (AUD) (development, persistence and desistence) by studying influences: of other substances (Aim 1);of parental characteristics including rearing styles and psychopathology (Aim 2);sibling influences (Aim 3);and in the context of 2 prominent models of AD vulnerability. Our aims together with design elements will continue progress towards understanding the etiology and course of AUDs applicable to adolescents and young adults in ethnically diverse families.